--- Log opened Mon Aug 28 00:00:30 2017
20170828 00:06:13-!- amello [~amello@179.159.56.56] has quit [Quit: Leaving]
20170828 00:13:16-!- Greg-Boggs [~greg_bogg@c-73-96-181-202.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20170828 00:21:35-!- Bonobo [~Bonobo@203.220.138.162] has joined #wesnoth-dev
20170828 00:30:50-!- gfgtdf [~chatzilla@x4e368ac9.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 55.0.3/20170824053622]]
20170828 00:36:08-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20170828 01:29:44-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:1df0:7030:91b8:9f4e] has joined #wesnoth-dev
20170828 01:43:23< irker321> wesnoth: Celtic Minstrel wesnoth:master 01892b16c75c / src/units/filter.hpp: Fix build on MSVC 2013 https://github.com/wesnoth/wesnoth/commit/01892b16c75c36696fb8f9369d88217d0e4c90d7
20170828 01:43:25< irker321> wesnoth: Celtic Minstrel wesnoth:master 45c8c256cccb / src/hotkey/ (command_executor.cpp hotkey_item.cpp): fix #1893 again (patch by @aginor) https://github.com/wesnoth/wesnoth/commit/45c8c256cccb299eab861d4e6f5712b72f200ada
20170828 01:44:16< celticminstrel> Sorry I didn't get that in earlier, zookeeper, I kinda got distracted.
20170828 01:47:57< celticminstrel> I guess I still need to figure out a solution to the scrolling hotkeys, but maybe that can wait for 1.13.10 or something.
20170828 01:55:21< irker321> wesnoth: Ignacio R. Morelle wesmere:develop 5f21f0036daf / wesmere/js/wiki.js: mw: Ensure dropdowns are at least as wide as their parent button https://github.com/wesnoth/wesmere/commit/5f21f0036daf1d86ac9e2ed8ee7c266f7e712973
20170828 02:24:38-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170828 02:24:44-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170828 02:31:31< irker321> wesnoth: Charles Dang wesnoth:master 8c8f109c5610 / src/map/ (label.cpp label.hpp): Map/Label: use default move ctor on compilers other than MSVC 2013 https://github.com/wesnoth/wesnoth/commit/8c8f109c56107e4b728f8abd5ba668b374f57f67
20170828 02:32:25< vultraz_iOS> celticminstrel: hotkeys are working again
20170828 02:53:19< celticminstrel> Well, I did apply Aginor's patch, so...
20170828 02:53:48-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has joined #wesnoth-dev
20170828 03:00:04-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:1df0:7030:91b8:9f4e] has quit [Remote host closed the connection]
20170828 03:40:00-!- JyrkiVesterinen [~jyrki@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170828 03:40:55< JyrkiVesterinen> Why does the unit_filter class have a move constructor?
20170828 03:41:10< JyrkiVesterinen> The move constructor copies. As far as I can tell, there is no performance advantage.
20170828 03:42:06-!- Greg-Boggs [~greg_bogg@c-73-96-181-202.hsd1.or.comcast.net] has joined #wesnoth-dev
20170828 03:42:20< celticminstrel> Hmm?
20170828 03:42:34< celticminstrel> Theoretically (assuming it's implemented properly) there should be an advantage if you have a large config?
20170828 03:42:42-!- Greg-Boggs [~greg_bogg@c-73-96-181-202.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20170828 03:42:46< celticminstrel> Since that would be moved instead of copied.
20170828 03:42:56< celticminstrel> Of course that's also assuming config itself has a move constructor though.
20170828 03:43:02< JyrkiVesterinen> Are configs movable?
20170828 03:43:35< celticminstrel> I dunno. They probably should be.
20170828 03:45:19< JyrkiVesterinen> Hmm, OK, they are indeed movable.
20170828 03:45:19< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/blob/01892b16c75c36696fb8f9369d88217d0e4c90d7/src/config.hpp#L117-L118
20170828 03:45:19< JyrkiVesterinen> So that explains the reason. Thanks. :)
20170828 03:45:51< celticminstrel> Even has separate move and copy assignment operators, huh...
20170828 03:46:19< celticminstrel> I was reading something somewhere that recommended not having those and instead defining T& operator=(T).
20170828 03:46:51< JyrkiVesterinen> I'd be interested to see that article.
20170828 03:46:56< celticminstrel> I guess it doesn't matter that much, though.
20170828 03:47:07< celticminstrel> It was long enough ago that it's probably purged from my history now...
20170828 03:47:14< celticminstrel> (browser history)
20170828 03:47:39< celticminstrel> Though I don't remember exactly how long, probably less than a year.
20170828 03:47:58< celticminstrel> I dunno, maybe I can find it again on google or something.
20170828 03:49:02< celticminstrel> Anyway it's basically the copy-and-swap idiom.
20170828 03:49:29< celticminstrel> Your assignment operator is literally T& operator=(T other) {swap(*this, other); return *this;}
20170828 03:50:56< JyrkiVesterinen> And the swap function has a moving rvalue reference version? Does it need to be implemented separately?
20170828 03:56:09< irker321> wesnoth: Jyrki Vesterinen wesnoth:master 9324349cb36d / src/map/ (label.cpp label.hpp): Revert "Map/Label: use default move ctor on compilers other than MSVC 2013" https://github.com/wesnoth/wesnoth/commit/9324349cb36df5d33843929f1fd193ffeccef247
20170828 03:56:42< JyrkiVesterinen> vultraz_iOS: Move constructors aren't magic pixie dust you can just add to make types movable.
20170828 03:56:51< JyrkiVesterinen> You need to understand what they actually do.
20170828 04:07:34< celticminstrel> The swap function does need to be implemented separately.
20170828 04:08:07< celticminstrel> Swap is inherently a move operation, so it takes non-const references.
20170828 04:14:19< JyrkiVesterinen> Oh, right.
20170828 04:15:20< JyrkiVesterinen> But I think you can't use swap to implement the *copy* constructor, when I think about it.
20170828 04:15:26< celticminstrel> I've also seen it recommended to call swap unqualified (ie, not std::swap)... something I haven't put into practice.
20170828 04:15:32< JyrkiVesterinen> The copy constructor can't alter the original object.
20170828 04:15:42< celticminstrel> This is for the assignment operators.
20170828 04:15:54< celticminstrel> You still need the separate move and copy constructors.
20170828 04:15:59< JyrkiVesterinen> Ah, right. Wrong word.
20170828 04:16:09< JyrkiVesterinen> But it doesn't change my point as far as I can tell.
20170828 04:16:22< JyrkiVesterinen> Copy assignment operator isn't allowed to alter the original object either.
20170828 04:16:58< celticminstrel> By original object you mean the object being copied, right?
20170828 04:17:07< celticminstrel> The argument to the assignment operator?
20170828 04:17:38< JyrkiVesterinen> Yes.
20170828 04:18:11< celticminstrel> That's why the argument is by value, not by reference.
20170828 04:18:19< celticminstrel> So that's the "copy" part of copy-and-swap.
20170828 04:18:43< JyrkiVesterinen> Hmm, I see. Indeed, it copies and then swaps.
20170828 04:19:39< JyrkiVesterinen> It's slower than only copying, though.
20170828 04:20:35< celticminstrel> I guess so. I think part of the big reason for it is exception-safety.
20170828 04:20:57-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Quit: Leaving]
20170828 04:45:03-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20170828 04:49:25-!- Greg-Boggs [~greg_bogg@c-73-96-181-202.hsd1.or.comcast.net] has joined #wesnoth-dev
20170828 05:23:39-!- Greg-Boggs [~greg_bogg@c-73-96-181-202.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20170828 05:32:14-!- JyrkiVesterinen [~jyrki@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170828 06:13:19-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170828 06:25:06-!- JyrkiVesterinen [~JyrkiVest@195-192-251-124.s1networks.fi] has joined #wesnoth-dev
20170828 06:47:32< zookeeper> Undefined macro in #ifver/#ifnver first argument: 'WESNOTH_VERSION'    at ~add-ons/Dragon_Lair/_main.cfg:35
20170828 06:47:41< zookeeper> the line being just this: #ifver WESNOTH_VERSION >= 1.13.2
20170828 06:47:53< zookeeper> didn't do anything with my previous month-old build, but latest master complains about that
20170828 06:49:59< shadowm> Force a cache refresh?
20170828 06:51:39< zookeeper> huh. yeah, i deleted the cache dir, that helped.
20170828 06:52:08< shadowm> It's a bug I've been observing since as far back as 1.9.x but I've never gotten around to properly documenting it in a bug report. If a WML read fails due to a parser or preprocessor error, the cache gets poisoned with an incomplete preprocessor configuration unit for that tree.
20170828 06:52:25< zookeeper> cool
20170828 06:52:56< shadowm> (I did intend to fix it myself before quitting, that's the primary reason why I never reported it.)
20170828 06:56:55-!- irker321 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170828 07:04:56< vultraz_iOS> shadowm: filed a bug
20170828 07:05:29< shadowm> Who where when why
20170828 07:05:45< vultraz_iOS> Me, GitHub, now, for the record
20170828 07:05:50< shadowm> About what?
20170828 07:05:55< vultraz_iOS> that
20170828 07:06:22< vultraz_iOS> of which you just spoke
20170828 07:06:24< shadowm> Was your IRC client designed by someone on drugs?
20170828 07:06:40< vultraz_iOS> maybe
20170828 07:06:59< shadowm> How does "Z" or "S" supposed to mean anything to anyone when there are a million people with nicknames that also start with any given letter of the Latin alphabet?
20170828 07:07:03< shadowm> *How is
20170828 07:07:35< vultraz_iOS> ...what the hell
20170828 07:07:48< vultraz_iOS> oh
20170828 07:07:50< vultraz_iOS> LOL
20170828 07:07:52< vultraz_iOS> wow
20170828 07:08:12< vultraz_iOS> ok so my IRC client has little colored circles next to names
20170828 07:08:16< vultraz_iOS> you are S
20170828 07:08:17< vultraz_iOS> mine is V
20170828 07:08:21< vultraz_iOS> zookeeper's is Z
20170828 07:08:23< vultraz_iOS> for obvious reasons
20170828 07:08:28-!- shadowm is now known as zhadowm
20170828 07:08:35< zhadowm> Hi.
20170828 07:08:41< vultraz_iOS> now it's Z
20170828 07:08:48< vultraz_iOS> and github collapses <>
20170828 07:08:49< zhadowm> How do you tell us apart now?
20170828 07:08:59< zookeeper> if you go to z'ha'dowm you will die
20170828 07:09:31< vultraz_iOS> zookeeper: your name is olive green and his is green green
20170828 07:09:33< vultraz_iOS> er
20170828 07:09:33< vultraz_iOS> no
20170828 07:09:38< vultraz_iOS> zhadowm: other way around
20170828 07:09:41< vultraz_iOS> YOU'RE olive green
20170828 07:09:49< zhadowm> You're an olive.
20170828 07:09:51-!- zhadowm is now known as shadowm
20170828 07:09:54< shadowm> Also, out of curiosity, any IRC clients for Android anyone would recommend?
20170828 07:10:18< shadowm> Wait.
20170828 07:10:22< shadowm> I mean SSH clients, not IRC clients.
20170828 07:11:55< vultraz_iOS> ssh has clients?
20170828 07:12:06< shadowm> ...
20170828 07:12:23< shadowm> vultraz, that's the weirdest question you've ever asked and that's saying a lot.
20170828 07:12:40< shadowm> It's almost like you have no idea what ssh even is in the first place.
20170828 07:12:51< vultraz_iOS> I thought it was a protocol 
20170828 07:12:56< shadowm> Yes, vultraz.
20170828 07:13:00< shadowm> IRC is a protocol.
20170828 07:13:02< shadowm> HTTP is a protocol.
20170828 07:13:17< shadowm> wesnothd's WML protocol is a protocol.
20170828 07:13:37< shadowm> You have a server and a client talking the protocol in question.
20170828 07:14:36< shadowm> In order for the protocol to be useful for anything at all you need a server and a client (which may or may not be the same entity, depending on the application).
20170828 07:17:58< vultraz_iOS> I see
20170828 07:27:16-!- atarocch [~atarocch@46.29.219.220] has joined #wesnoth-dev
20170828 08:40:32-!- Duthlet [~Duthlet@ipservice-092-211-168-098.092.211.pools.vodafone-ip.de] has joined #wesnoth-dev
20170828 09:01:47< loonycyborg> shadowm: I'm using connectbot
20170828 09:02:46< loonycyborg> as ssh client on android that is :P
20170828 09:49:23-!- irker048 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170828 09:49:23< irker048> wesnoth: Charles Dang wesnoth:master 699047766a88 / src/ (3 files in 2 dirs): Refactor out wesnothd_connection_ptr wrapper class https://github.com/wesnoth/wesnoth/commit/699047766a8852014df20c44d5ed84dadb37015e
20170828 10:48:09< irker048> wesnoth: Charles Dang wesnoth:master 520371bc105f / src/gui/ (dialogs/game_load.cpp widgets/text_box_base.hpp): GUI2/Game Load: don't delete saves if filter text box has focus! https://github.com/wesnoth/wesnoth/commit/520371bc105f62d39ace4237ebdaa2172a6d95a4
20170828 11:24:37< zookeeper> if a WML key can be passed either a bool or a positive int, and the value is mapped internally to an int (true == -1 and false == 0) into random_start_i, then would this be an acceptable manner of doing it?
20170828 11:24:38< zookeeper> int random_start_i = img["random_start"].to_int(-1);
20170828 11:24:38< zookeeper> bool random_start_b = img["random_start"].to_bool(true);
20170828 11:24:38< zookeeper> if (!random_start_b) random_start_i = 0;
20170828 11:24:44< zookeeper> ...it seems to work correctly, anyway
20170828 11:25:05< vultraz_iOS> uh
20170828 11:25:08< vultraz_iOS> what are you trying to do?
20170828 11:25:18< zookeeper> <zookeeper> if a WML key can be passed either a bool or a positive int
20170828 11:25:33< JyrkiVesterinen> vultraz_iOS, you have seen our discussion in Discord, right?
20170828 11:25:52< vultraz_iOS> i have not followed it closely 
20170828 11:27:22< JyrkiVesterinen> The point is introducing some variation to water animation start times so that every single water hex doesn't redraw at once.
20170828 11:27:44< JyrkiVesterinen> By allowing random_start= to accept both boolean and numeric values.
20170828 11:27:48< zookeeper> of course i could just add a new key instead, but that's more clunky and verbose, whereas the semantics of overloading random_start= seem pretty clear to me
20170828 12:05:10< irker048> wesnoth: Charles Dang wesnoth:master d3fe82bed8d6 / src/game_initialization/ (lobby_info.cpp lobby_info.hpp): Game Initialization/Lobby Info: use unique_ptr for game_info map https://github.com/wesnoth/wesnoth/commit/d3fe82bed8d6f649fb4759746121e5462fc2b35d
20170828 12:26:33< Soliton> zookeeper: i'd do: int random_start_i = random_start_b ? -1 : 0;
20170828 12:27:26< zookeeper> but that only assigns -1 or 0
20170828 12:27:49< Soliton> that's how i understood the task.
20170828 12:30:59< zookeeper> random_start=yes -> -1   |   random_start=no -> 0   |   random_start=7 -> 7
20170828 12:31:54< Soliton> random_start=-100 -> ?
20170828 12:32:00< zookeeper> *explosion*
20170828 12:32:04< Soliton> cool!
20170828 12:32:14< zookeeper> (-1 or 0, whatever)
20170828 12:33:02< Soliton> i'd guess it'll be -100 with your code. haven't checked in detail though.
20170828 12:33:44< zookeeper> ah, true. well, still wouldn't break anything.
20170828 12:34:15< zookeeper> the code would just check for < 0 later so doesn't matter which negative int it is
20170828 12:35:40< JyrkiVesterinen> It might be better to throw a WML exception (which results in an error message).
20170828 12:36:19< JyrkiVesterinen> In other software projects there have been too many cases where developers have relied on undocumented behavior and changes have been impossible afterwards.
20170828 12:45:03-!- travis-ci [~travis-ci@ec2-174-129-96-232.compute-1.amazonaws.com] has joined #wesnoth-dev
20170828 12:45:04< travis-ci> wesnoth/wesnoth#14819 (master - d3fe82b : Charles Dang): The build was broken.
20170828 12:45:04< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/269166139
20170828 12:45:04-!- travis-ci [~travis-ci@ec2-174-129-96-232.compute-1.amazonaws.com] has left #wesnoth-dev []
20170828 12:56:12-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170828 12:56:19-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170828 13:07:08-!- gfgtdf [~chatzilla@x4e3635c1.dyn.telefonica.de] has joined #wesnoth-dev
20170828 13:08:24< gfgtdf> vultraz_iOS: why did you remove the netowrk_connection_ptr ?
20170828 13:08:36< vultraz_iOS> gfgtdf: wasn't necessary
20170828 13:08:56< gfgtdf> vultraz_iOS: it ewas it's pupose was to call stop() when the pointer is destructed, 
20170828 13:09:19< vultraz_iOS> gfgtdf: i moved that call to the wesnothd_connection dtor
20170828 13:09:28< gfgtdf> vultraz_iOS: so that even if there is a second, shared_ptr poining to it, it will still be destructed.
20170828 13:09:39< gfgtdf> it will still call ::stop i mean
20170828 13:10:33< gfgtdf> vultraz_iOS: that's also why the old connection_pt class had "wesnothd_connection_ptr(const wesnothd_connection_ptr&) = delete;"
20170828 13:11:06< vultraz_iOS> you could make it a unique_ptr
20170828 13:12:10< gfgtdf> thn it wouldn't be shared, as i said before it there are multiple shared_ptrs pointing to this object
20170828 13:16:42< gfgtdf> s/before it there/before there 
20170828 13:27:46< gfgtdf> vultraz_iOS: please revert it (or proove me that it still works in case that there is something i didn't see).
20170828 13:28:14< vultraz_iOS> working on something else r/n will get to it
20170828 13:36:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170828 13:36:41-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170828 13:37:50< vultraz_iOS> gfgtdf: everything builds fine if wesnothd_connection_ptr is a unique_ptr
20170828 13:38:35< gfgtdf> but then you have a shared_ptr and a unitque_ptr pointing to the same object which coudl be troublesome
20170828 13:38:43< vultraz_iOS> what?
20170828 13:38:50< gfgtdf> vultraz_iOS: the way it curretlly (before your commit) worked is tis:
20170828 13:39:05< gfgtdf> vultraz_iOS: there are at most 2 shaed_ptrs poinint to the connection object:
20170828 13:39:30< gfgtdf> vultraz_iOS: this one: https://github.com/wesnoth/wesnoth/blob/2eaf65bfb087fcfb47a4e0c3fef166d93324239a/src/wesnothd_connection.hpp#L206
20170828 13:39:43< gfgtdf> vultraz_iOS: and this one: https://github.com/wesnoth/wesnoth/blob/2eaf65bfb087fcfb47a4e0c3fef166d93324239a/src/wesnothd_connection.cpp#L153
20170828 13:40:13< gfgtdf> (the second oen is inside the workder thread, to amke sure the connection object isn't destucted before the worker threa had finished)
20170828 13:40:16< gfgtdf> make*
20170828 13:41:57< gfgtdf> so what shodul happen when the wesnotd_conetion_ptr is destucted is that: 1) it calls connection::stop(), 2) that end the worker thread, 3) so the shared_ptr in the worker thread  is destructed, 4) the object id feed with the last shared ptr pointing to it beeing destructed
20170828 13:42:24< gfgtdf> ends*
20170828 13:42:34< gfgtdf> s/feed/freed*
20170828 13:44:14< vultraz_iOS> ok, so what's the problem after my commit
20170828 13:47:42< gfgtdf> vultraz_iOS: that stop() isn'T called after the wesntohd_connection_ptr is destructed (the shared_ptr at wesnothd_connection.cpp#L153 still point to it)
20170828 13:47:46< gfgtdf> isn't
20170828 13:49:26< vultraz_iOS> it really seems there should be a cleaner way to just ensure the object isn't destroyed before the worker thread terminates
20170828 13:51:41< vultraz_iOS> (plus the smart pointers can take custom deleters so that could also be used somehow)
20170828 13:58:42< gfgtdf> the custom deleters do not change when the deleter is called, there is also the possibelity tht the wesntohd_connection_ptr is destructed before the popinte at wesnothd_connection.cpp#L153  is created, in which case you want to destuct the object right away, so i do think a cusotm class that used shared_ptr internally is the simplest way to solve it.
20170828 13:59:23< vultraz_iOS> >_>
20170828 13:59:35-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:a414:a794:595f:c769] has joined #wesnoth-dev
20170828 14:00:31< gfgtdf> s/popinte/pointer
20170828 14:17:57-!- Kwandulin [~Kwandulin@p200300E453CC3B7E5D7615C11F82772C.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170828 14:20:59-!- JyrkiVesterinen [~JyrkiVest@195-192-251-124.s1networks.fi] has quit [Quit: .]
20170828 14:40:34< vultraz_iOS> gfgtdf: btw if halt() is supposed to be called when the ptr is destroyed why was it in the assignment operator
20170828 14:40:36-!- Greg-Boggs [~greg_bogg@2601:1c2:f00:9780:a414:a794:595f:c769] has quit [Remote host closed the connection]
20170828 14:40:49< gfgtdf> you mean stop() ?
20170828 14:40:58< vultraz_iOS> Yes
20170828 14:41:35< gfgtdf> because the asignement operator usually replaced 'this' with the other object, which 'destroys' the previosuly value of this
20170828 14:41:46< gfgtdf> previous*
20170828 14:44:41< vultraz_iOS> Right but that means if the ptr wrapper was only destroyed stop wouldn't get called
20170828 14:51:48< gfgtdf> vultraz_iOS: not sure what you mean here, both destructor and the assignment oprereator call stop() because both remvoe the pointer to the connection object
20170828 14:53:02< vultraz_iOS> Oh sorry didn't see that part
20170828 14:53:09-!- Duthlet [~Duthlet@ipservice-092-211-168-098.092.211.pools.vodafone-ip.de] has quit [Ping timeout: 252 seconds]
20170828 15:05:42-!- irker048 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170828 15:06:15< gfgtdf> in fact i think currently the operator= is mostly used as "ptr = wesnothd_connection_ptr()" to reset the object.
20170828 15:31:20-!- irker847 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170828 15:31:20< irker847> wesnoth: Charles Dang wesnoth:preprocessor_cleanup e6b5b260b25f / src/serialization/ (preprocessor.cpp preprocessor.hpp): Preprocessor: reformat this absolute mess of a file https://github.com/wesnoth/wesnoth/commit/e6b5b260b25fd01e5fbce9d1d32f8fc1d5640f93
20170828 15:31:20< irker847> wesnoth: Charles Dang wesnoth:preprocessor_cleanup 5ab2ec5b5829 / src/serialization/preprocessor.cpp: Preprocessor: used emplace_back in one place https://github.com/wesnoth/wesnoth/commit/5ab2ec5b582965d81d4f3668a0c1c2a0b40ad70d
20170828 15:31:21< irker847> wesnoth: Charles Dang wesnoth:preprocessor_cleanup a204a87ba54f / src/serialization/preprocessor.cpp: Preprocessor: use more logical method to assign current streambuf target https://github.com/wesnoth/wesnoth/commit/a204a87ba54f0d7d410ada2cb71f39459accfc44
20170828 15:35:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 15:36:21-!- Bonobo [~Bonobo@203.220.138.162] has quit [Ping timeout: 240 seconds]
20170828 15:40:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170828 15:45:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 15:50:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20170828 16:02:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 16:22:37-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170828 16:52:15-!- travis-ci [~travis-ci@ec2-54-211-173-142.compute-1.amazonaws.com] has joined #wesnoth-dev
20170828 16:52:16< travis-ci> wesnoth/wesnoth#14820 (preprocessor_cleanup - a204a87 : Charles Dang): The build has errored.
20170828 16:52:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/269241503
20170828 16:52:16-!- travis-ci [~travis-ci@ec2-54-211-173-142.compute-1.amazonaws.com] has left #wesnoth-dev []
20170828 17:36:13-!- Kwandulin [~Kwandulin@p200300E453CC3B7E5D7615C11F82772C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170828 17:38:44< irker847> wesnoth: Jyrki Vesterinen wesnoth:master f730338222d7 / src/game_initialization/lobby_info.cpp: Fix build https://github.com/wesnoth/wesnoth/commit/f730338222d7d8842f3eda3b56fce55e9a8334b3
20170828 17:45:04-!- Appveyor [~Appveyor@67.225.139.254] has joined #wesnoth-dev
20170828 17:45:04< Appveyor> The Battle for Wesnoth (Visual Studio 2013) - Release Charles Dang d3fe82b: Game Initialization/Lobby Info: use unique_ptr for game_info map Failed
20170828 17:45:04< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth/build/Wesnoth-VS2013-master-445
20170828 17:45:08-!- Appveyor [~Appveyor@67.225.139.254] has left #wesnoth-dev []
20170828 17:58:39-!- Appveyor [~Appveyor@67.225.139.254] has joined #wesnoth-dev
20170828 17:58:39< Appveyor> The Battle for Wesnoth (Visual Studio 2013) - Debug Charles Dang d3fe82b: Game Initialization/Lobby Info: use unique_ptr for game_info map Failed
20170828 17:58:39< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth/build/Wesnoth-VS2013-master-445
20170828 17:58:43-!- Appveyor [~Appveyor@67.225.139.254] has left #wesnoth-dev []
20170828 18:02:32-!- mjs-de [~mjs-de@dorf.cc] has joined #wesnoth-dev
20170828 18:08:59-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170828 18:14:04-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev
20170828 18:26:35-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 240 seconds]
20170828 18:28:51-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20170828 18:31:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 18:37:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 18:52:15-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170828 19:14:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 19:26:40-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170828 19:54:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170828 20:00:14-!- gfgtdf [~chatzilla@x4e3635c1.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170828 20:01:33-!- mjs-de [~mjs-de@dorf.cc] has quit [Remote host closed the connection]
20170828 20:02:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 20:06:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 248 seconds]
20170828 20:17:18-!- gfgtdf [~chatzilla@x4e3635c1.dyn.telefonica.de] has joined #wesnoth-dev
20170828 20:18:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 20:39:22-!- irker847 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170828 20:44:29-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has quit [Remote host closed the connection]
20170828 21:09:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 21:16:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 21:24:12-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection]
20170828 22:26:51-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170828 22:26:58-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170828 22:30:10-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds]
20170828 22:44:10-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Ping timeout: 264 seconds]
20170828 22:45:21-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20170828 22:55:19-!- irker081 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170828 22:55:19< irker081> wesnoth: Ignacio R. Morelle wesnoth:master c0223cd7c0a7 / data/tools/unit_tree/html_output.py: wmlunits: Redo some bizarre bogus maths I did for the language menu https://github.com/wesnoth/wesnoth/commit/c0223cd7c0a7b4ecac457b1fc123ced8350a723a
20170828 22:59:24-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170828 23:07:50-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170828 23:32:05-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20170828 23:39:10-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20170828 23:49:16-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170828 23:50:36< celticminstrel> vultraz_iOS: So the first commit is strictly reformatting, nothing else?
20170828 23:51:25< vultraz_iOS> Did remove some forward declarations that didn't work but yeah
20170828 23:51:40< vultraz_iOS> ignore it
20170828 23:51:51< vultraz_iOS> im still pondering how to fix the last commit
20170828 23:52:06< celticminstrel> Guessing you're still working on it since there's barely anything there besides the reformatting.
20170828 23:52:12< vultraz_iOS> yes
20170828 23:52:41< vultraz_iOS> the problem is the ptr needs to be set *first*...
20170828 23:52:58< vultraz_iOS> yet it's confusing to have plain `new` statements since it looks like a memory leak
20170828 23:54:49< vultraz_iOS> basically the use of `new` seems to be so the objects don'tget destroyed
20170828 23:55:34< celticminstrel> It's kinda like one of the use-cases of a memory pool.
20170828 23:55:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 23:55:49< vultraz_iOS> awhat now?
20170828 23:55:50< celticminstrel> It wants to allocate a bunch of things but isn't sure how long it'll need them for.
20170828 23:56:01< celticminstrel> So it just allocates them and stores a reference in a list.
20170828 23:56:14< celticminstrel> Then when the preprocessing is done it deallocates them all at the same time.
20170828 23:56:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 23:56:17< celticminstrel> IIRC
20170828 23:56:28< celticminstrel> That's not the primary use-case of a memory pool, admittedly.
20170828 23:57:37-!- gfgtdf_ [~chatzilla@x4e3635c1.dyn.telefonica.de] has joined #wesnoth-dev
20170828 23:57:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 23:57:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 23:57:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170828 23:57:57< celticminstrel> If it were an actual memory pool you'd be allocating a large chunk up front and then portinoing that out to the various things you need.
20170828 23:58:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170828 23:58:32-!- gfgtdf [~chatzilla@x4e3635c1.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20170828 23:58:38< celticminstrel> Anyway, I think the point is that it doesn't know exactly how long the objects need to remain valid.
20170828 23:58:47-!- gfgtdf_ is now known as gfgtdf
--- Log closed Tue Aug 29 00:00:03 2017